[PAtCh 0/2] cgroup: prepare kernel config for cgroup2

[PAtCh 0/2] cgroup: prepare kernel config for cgroup2

[Johannes Weiner]

For the scheduled release of cgroup2, put the cgroup Kconfig section
in order. Fixes the names of the controller items, then puts them in
order that makes it easy for people to quickly find what they need:
core features first, then more obscure and legacy stuff at the end.

[PATCH 1/2] cgroup: clean up the kernel configuration menu nomenclature

[Johannes Weiner]

The config options for the different cgroup controllers use various
terms: resource controller, cgroup subsystem, etc. Simplify this to
"controller", which is clear enough in the cgroup context.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
---
 init/Kconfig | 65 ++++++++++++++++++++++++------------------------------------
 1 file changed, 26 insertions(+), 39 deletions(-)

diff --git a/init/Kconfig b/init/Kconfig
index 16840b4..ed324f5 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -958,23 +958,22 @@ menuconfig CGROUPS
 if CGROUPS
 
 config CGROUP_DEBUG
-	bool "Example debug cgroup subsystem"
+	bool "Example controller"
 	default n
 	help
-	  This option enables a simple cgroup subsystem that
-	  exports useful debugging information about the cgroups
-	  framework.
+	  This option enables a simple controller that exports
+	  debugging information about the cgroups framework.
 
-	  Say N if unsure.
+	  Say N.
 
 config CGROUP_FREEZER
-	bool "Freezer cgroup subsystem"
+	bool "Freezer controller"
 	help
 	  Provides a way to freeze and unfreeze all tasks in a
 	  cgroup.
 
 config CGROUP_PIDS
-	bool "PIDs cgroup subsystem"
+	bool "PIDs controller"
 	help
 	  Provides enforcement of process number limits in the scope of a
 	  cgroup. Any attempt to fork more processes than is allowed in the
@@ -990,13 +989,13 @@ config CGROUP_PIDS
 	  attach to a cgroup.
 
 config CGROUP_DEVICE
-	bool "Device controller for cgroups"
+	bool "Device controller"
 	help
-	  Provides a cgroup implementing whitelists for devices which
-	  a process in the cgroup can mknod or open.
+	  Provides a cgroup controller implementing whitelists for
+	  devices which a process in the cgroup can mknod or open.
 
 config CPUSETS
-	bool "Cpuset support"
+	bool "Cpuset controller"
 	help
 	  This option will let you create and manage CPUSETs which
 	  allow dynamically partitioning a system into sets of CPUs and
@@ -1011,41 +1010,29 @@ config PROC_PID_CPUSET
 	default y
 
 config CGROUP_CPUACCT
-	bool "Simple CPU accounting cgroup subsystem"
+	bool "Simple CPU accounting controller"
 	help
-	  Provides a simple Resource Controller for monitoring the
+	  Provides a simple controller for monitoring the
 	  total CPU consumed by the tasks in a cgroup.
 
 config PAGE_COUNTER
        bool
 
 config MEMCG
-	bool "Memory Resource Controller for Control Groups"
+	bool "Memory controller"
 	select PAGE_COUNTER
 	select EVENTFD
 	help
-	  Provides a memory resource controller that manages both anonymous
-	  memory and page cache. (See Documentation/cgroups/memory.txt)
+	  Provides control over the memory footprint of tasks in a cgroup.
 
 config MEMCG_SWAP
-	bool "Memory Resource Controller Swap Extension"
+	bool "Swap controller"
 	depends on MEMCG && SWAP
 	help
-	  Add swap management feature to memory resource controller. When you
-	  enable this, you can limit mem+swap usage per cgroup. In other words,
-	  when you disable this, memory resource controller has no cares to
-	  usage of swap...a process can exhaust all of the swap. This extension
-	  is useful when you want to avoid exhaustion swap but this itself
-	  adds more overheads and consumes memory for remembering information.
-	  Especially if you use 32bit system or small memory system, please
-	  be careful about enabling this. When memory resource controller
-	  is disabled by boot option, this will be automatically disabled and
-	  there will be no overhead from this. Even when you set this config=y,
-	  if boot option "swapaccount=0" is set, swap will not be accounted.
-	  Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
-	  size is 4096bytes, 512k per 1Gbytes of swap.
+	  Provides control over the swap space consumed by tasks in a cgroup.
+
 config MEMCG_SWAP_ENABLED
-	bool "Memory Resource Controller Swap Extension enabled by default"
+	bool "Swap controller enabled by default"
 	depends on MEMCG_SWAP
 	default y
 	help
@@ -1059,12 +1046,12 @@ config MEMCG_SWAP_ENABLED
 	  then swapaccount=0 does the trick).
 
 config CGROUP_HUGETLB
-	bool "HugeTLB Resource Controller for Control Groups"
+	bool "HugeTLB controller"
 	depends on HUGETLB_PAGE
 	select PAGE_COUNTER
 	default n
 	help
-	  Provides a cgroup Resource Controller for HugeTLB pages.
+	  Provides a cgroup controller for HugeTLB pages.
 	  When you enable this, you can put a per cgroup limit on HugeTLB usage.
 	  The limit is enforced during page fault. Since HugeTLB doesn't
 	  support page reclaim, enforcing the limit at page fault time implies
@@ -1075,17 +1062,17 @@ config CGROUP_HUGETLB
 	  that we cannot use the controller with huge page less than 3 pages.
 
 config CGROUP_PERF
-	bool "Enable perf_event per-cpu per-container group (cgroup) monitoring"
+	bool "Perf controller"
 	depends on PERF_EVENTS && CGROUPS
 	help
-	  This option extends the per-cpu mode to restrict monitoring to
-	  threads which belong to the cgroup specified and run on the
+	  This option extends the perf per-cpu mode to restrict monitoring
+	  to threads which belong to the cgroup specified and run on the
 	  designated cpu.
 
 	  Say N if unsure.
 
 menuconfig CGROUP_SCHED
-	bool "Group CPU scheduler"
+	bool "CPU controller"
 	default n
 	help
 	  This feature lets CPU scheduler recognize task groups and control CPU
@@ -1123,7 +1110,7 @@ config RT_GROUP_SCHED
 endif #CGROUP_SCHED
 
 config BLK_CGROUP
-	bool "Block IO controller"
+	bool "IO controller"
 	depends on BLOCK
 	default n
 	---help---
@@ -1145,7 +1132,7 @@ config BLK_CGROUP
 	See Documentation/cgroups/blkio-controller.txt for more information.
 
 config DEBUG_BLK_CGROUP
-	bool "Enable Block IO controller debugging"
+	bool "IO controller debugging"
 	depends on BLK_CGROUP
 	default n
 	---help---
-- 
2.6.4

[PATCH 2/2] cgroup: put controller Kconfig options in meaningful order

[Johannes Weiner]

To make it easier to quickly find what's needed list the basic
resource controllers of cgroup2 first - io, memory, cpu - while
pushing the more exotic and/or legacy controllers to the bottom.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
---
 init/Kconfig | 214 +++++++++++++++++++++++++++++------------------------------
 1 file changed, 107 insertions(+), 107 deletions(-)

diff --git a/init/Kconfig b/init/Kconfig
index ed324f5..fdb5ecb 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -957,64 +957,6 @@ menuconfig CGROUPS
 
 if CGROUPS
 
-config CGROUP_DEBUG
-	bool "Example controller"
-	default n
-	help
-	  This option enables a simple controller that exports
-	  debugging information about the cgroups framework.
-
-	  Say N.
-
-config CGROUP_FREEZER
-	bool "Freezer controller"
-	help
-	  Provides a way to freeze and unfreeze all tasks in a
-	  cgroup.
-
-config CGROUP_PIDS
-	bool "PIDs controller"
-	help
-	  Provides enforcement of process number limits in the scope of a
-	  cgroup. Any attempt to fork more processes than is allowed in the
-	  cgroup will fail. PIDs are fundamentally a global resource because it
-	  is fairly trivial to reach PID exhaustion before you reach even a
-	  conservative kmemcg limit. As a result, it is possible to grind a
-	  system to halt without being limited by other cgroup policies. The
-	  PIDs cgroup subsystem is designed to stop this from happening.
-
-	  It should be noted that organisational operations (such as attaching
-	  to a cgroup hierarchy will *not* be blocked by the PIDs subsystem),
-	  since the PIDs limit only affects a process's ability to fork, not to
-	  attach to a cgroup.
-
-config CGROUP_DEVICE
-	bool "Device controller"
-	help
-	  Provides a cgroup controller implementing whitelists for
-	  devices which a process in the cgroup can mknod or open.
-
-config CPUSETS
-	bool "Cpuset controller"
-	help
-	  This option will let you create and manage CPUSETs which
-	  allow dynamically partitioning a system into sets of CPUs and
-	  Memory Nodes and assigning tasks to run only within those sets.
-	  This is primarily useful on large SMP or NUMA systems.
-
-	  Say N if unsure.
-
-config PROC_PID_CPUSET
-	bool "Include legacy /proc/<pid>/cpuset file"
-	depends on CPUSETS
-	default y
-
-config CGROUP_CPUACCT
-	bool "Simple CPU accounting controller"
-	help
-	  Provides a simple controller for monitoring the
-	  total CPU consumed by the tasks in a cgroup.
-
 config PAGE_COUNTER
        bool
 
@@ -1045,31 +987,40 @@ config MEMCG_SWAP_ENABLED
 	  select this option (if, for some reason, they need to disable it
 	  then swapaccount=0 does the trick).
 
-config CGROUP_HUGETLB
-	bool "HugeTLB controller"
-	depends on HUGETLB_PAGE
-	select PAGE_COUNTER
+config BLK_CGROUP
+	bool "IO controller"
+	depends on BLOCK
 	default n
-	help
-	  Provides a cgroup controller for HugeTLB pages.
-	  When you enable this, you can put a per cgroup limit on HugeTLB usage.
-	  The limit is enforced during page fault. Since HugeTLB doesn't
-	  support page reclaim, enforcing the limit at page fault time implies
-	  that, the application will get SIGBUS signal if it tries to access
-	  HugeTLB pages beyond its limit. This requires the application to know
-	  beforehand how much HugeTLB pages it would require for its use. The
-	  control group is tracked in the third page lru pointer. This means
-	  that we cannot use the controller with huge page less than 3 pages.
+	---help---
+	Generic block IO controller cgroup interface. This is the common
+	cgroup interface which should be used by various IO controlling
+	policies.
 
-config CGROUP_PERF
-	bool "Perf controller"
-	depends on PERF_EVENTS && CGROUPS
-	help
-	  This option extends the perf per-cpu mode to restrict monitoring
-	  to threads which belong to the cgroup specified and run on the
-	  designated cpu.
+	Currently, CFQ IO scheduler uses it to recognize task groups and
+	control disk bandwidth allocation (proportional time slice allocation)
+	to such task groups. It is also used by bio throttling logic in
+	block layer to implement upper limit in IO rates on a device.
 
-	  Say N if unsure.
+	This option only enables generic Block IO controller infrastructure.
+	One needs to also enable actual IO controlling logic/policy. For
+	enabling proportional weight division of disk bandwidth in CFQ, set
+	CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
+	CONFIG_BLK_DEV_THROTTLING=y.
+
+	See Documentation/cgroups/blkio-controller.txt for more information.
+
+config DEBUG_BLK_CGROUP
+	bool "IO controller debugging"
+	depends on BLK_CGROUP
+	default n
+	---help---
+	Enable some debugging help. Currently it exports additional stat
+	files in a cgroup which can be useful for debugging.
+
+config CGROUP_WRITEBACK
+	bool
+	depends on MEMCG && BLK_CGROUP
+	default y
 
 menuconfig CGROUP_SCHED
 	bool "CPU controller"
@@ -1109,40 +1060,89 @@ config RT_GROUP_SCHED
 
 endif #CGROUP_SCHED
 
-config BLK_CGROUP
-	bool "IO controller"
-	depends on BLOCK
+config CGROUP_PIDS
+	bool "PIDs controller"
+	help
+	  Provides enforcement of process number limits in the scope of a
+	  cgroup. Any attempt to fork more processes than is allowed in the
+	  cgroup will fail. PIDs are fundamentally a global resource because it
+	  is fairly trivial to reach PID exhaustion before you reach even a
+	  conservative kmemcg limit. As a result, it is possible to grind a
+	  system to halt without being limited by other cgroup policies. The
+	  PIDs cgroup subsystem is designed to stop this from happening.
+
+	  It should be noted that organisational operations (such as attaching
+	  to a cgroup hierarchy will *not* be blocked by the PIDs subsystem),
+	  since the PIDs limit only affects a process's ability to fork, not to
+	  attach to a cgroup.
+
+config CGROUP_FREEZER
+	bool "Freezer controller"
+	help
+	  Provides a way to freeze and unfreeze all tasks in a
+	  cgroup.
+
+config CGROUP_HUGETLB
+	bool "HugeTLB controller"
+	depends on HUGETLB_PAGE
+	select PAGE_COUNTER
 	default n
-	---help---
-	Generic block IO controller cgroup interface. This is the common
-	cgroup interface which should be used by various IO controlling
-	policies.
+	help
+	  Provides a cgroup controller for HugeTLB pages.
+	  When you enable this, you can put a per cgroup limit on HugeTLB usage.
+	  The limit is enforced during page fault. Since HugeTLB doesn't
+	  support page reclaim, enforcing the limit at page fault time implies
+	  that, the application will get SIGBUS signal if it tries to access
+	  HugeTLB pages beyond its limit. This requires the application to know
+	  beforehand how much HugeTLB pages it would require for its use. The
+	  control group is tracked in the third page lru pointer. This means
+	  that we cannot use the controller with huge page less than 3 pages.
 
-	Currently, CFQ IO scheduler uses it to recognize task groups and
-	control disk bandwidth allocation (proportional time slice allocation)
-	to such task groups. It is also used by bio throttling logic in
-	block layer to implement upper limit in IO rates on a device.
+config CPUSETS
+	bool "Cpuset controller"
+	help
+	  This option will let you create and manage CPUSETs which
+	  allow dynamically partitioning a system into sets of CPUs and
+	  Memory Nodes and assigning tasks to run only within those sets.
+	  This is primarily useful on large SMP or NUMA systems.
 
-	This option only enables generic Block IO controller infrastructure.
-	One needs to also enable actual IO controlling logic/policy. For
-	enabling proportional weight division of disk bandwidth in CFQ, set
-	CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
-	CONFIG_BLK_DEV_THROTTLING=y.
+	  Say N if unsure.
 
-	See Documentation/cgroups/blkio-controller.txt for more information.
+config PROC_PID_CPUSET
+	bool "Include legacy /proc/<pid>/cpuset file"
+	depends on CPUSETS
+	default y
 
-config DEBUG_BLK_CGROUP
-	bool "IO controller debugging"
-	depends on BLK_CGROUP
+config CGROUP_DEVICE
+	bool "Device controller"
+	help
+	  Provides a cgroup controller implementing whitelists for
+	  devices which a process in the cgroup can mknod or open.
+
+config CGROUP_CPUACCT
+	bool "Simple CPU accounting controller"
+	help
+	  Provides a simple controller for monitoring the
+	  total CPU consumed by the tasks in a cgroup.
+
+config CGROUP_PERF
+	bool "Perf controller"
+	depends on PERF_EVENTS && CGROUPS
+	help
+	  This option extends the perf per-cpu mode to restrict monitoring
+	  to threads which belong to the cgroup specified and run on the
+	  designated cpu.
+
+	  Say N if unsure.
+
+config CGROUP_DEBUG
+	bool "Example controller"
 	default n
-	---help---
-	Enable some debugging help. Currently it exports additional stat
-	files in a cgroup which can be useful for debugging.
+	help
+	  This option enables a simple controller that exports
+	  debugging information about the cgroups framework.
 
-config CGROUP_WRITEBACK
-	bool
-	depends on MEMCG && BLK_CGROUP
-	default y
+	  Say N.
 
 endif # CGROUPS
 
-- 
2.6.4

Re: [PATCH 1/2] cgroup: clean up the kernel configuration menu nomenclature

[Zefan Li]

On 2015/12/18 6:19, Johannes Weiner wrote:
> The config options for the different cgroup controllers use various
> terms: resource controller, cgroup subsystem, etc. Simplify this to
> "controller", which is clear enough in the cgroup context.
>
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> ---
>   init/Kconfig | 65 ++++++++++++++++++++++++------------------------------------
>   1 file changed, 26 insertions(+), 39 deletions(-)
>
> diff --git a/init/Kconfig b/init/Kconfig
> index 16840b4..ed324f5 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -958,23 +958,22 @@ menuconfig CGROUPS
>   if CGROUPS
>
>   config CGROUP_DEBUG
> -	bool "Example debug cgroup subsystem"
> +	bool "Example controller"

But not all cgroup subsystems are controllers, which are debug, perf and
cpuacct. I'm fine the other changes.

Re: [PATCH 2/2] cgroup: put controller Kconfig options in meaningful order

[Zefan Li]

On 2015/12/18 6:19, Johannes Weiner wrote:
> To make it easier to quickly find what's needed list the basic
> resource controllers of cgroup2 first - io, memory, cpu - while
> pushing the more exotic and/or legacy controllers to the bottom.
>
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>

Acked-by: Zefan Li <lizefan@huawei.com>

> ---
>   init/Kconfig | 214 +++++++++++++++++++++++++++++------------------------------
>   1 file changed, 107 insertions(+), 107 deletions(-)
>
> diff --git a/init/Kconfig b/init/Kconfig
> index ed324f5..fdb5ecb 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -957,64 +957,6 @@ menuconfig CGROUPS
>
...
> +config CGROUP_PERF
> +	bool "Perf controller"
> +	depends on PERF_EVENTS && CGROUPS

While you are at it, remove "&& CGROUPS" ?

> +	help
> +	  This option extends the perf per-cpu mode to restrict monitoring
> +	  to threads which belong to the cgroup specified and run on the
> +	  designated cpu.
> +
> +	  Say N if unsure.
> +
> +config CGROUP_DEBUG
> +	bool "Example controller"
>   	default n
> -	---help---
> -	Enable some debugging help. Currently it exports additional stat
> -	files in a cgroup which can be useful for debugging.
> +	help
> +	  This option enables a simple controller that exports
> +	  debugging information about the cgroups framework.
>
> -config CGROUP_WRITEBACK
> -	bool
> -	depends on MEMCG && BLK_CGROUP
> -	default y
> +	  Say N.
>
>   endif # CGROUPS
>
>

Re: [PATCH 1/2] cgroup: clean up the kernel configuration menu nomenclature

[Johannes Weiner]

On Fri, Dec 18, 2015 at 10:23:03AM +0800, Zefan Li wrote:
> On 2015/12/18 6:19, Johannes Weiner wrote:
> >The config options for the different cgroup controllers use various
> >terms: resource controller, cgroup subsystem, etc. Simplify this to
> >"controller", which is clear enough in the cgroup context.
> >
> >Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> >---
> >  init/Kconfig | 65 ++++++++++++++++++++++++------------------------------------
> >  1 file changed, 26 insertions(+), 39 deletions(-)
> >
> >diff --git a/init/Kconfig b/init/Kconfig
> >index 16840b4..ed324f5 100644
> >--- a/init/Kconfig
> >+++ b/init/Kconfig
> >@@ -958,23 +958,22 @@ menuconfig CGROUPS
> >  if CGROUPS
> >
> >  config CGROUP_DEBUG
> >-	bool "Example debug cgroup subsystem"
> >+	bool "Example controller"
> 
> But not all cgroup subsystems are controllers, which are debug, perf and
> cpuacct. I'm fine the other changes.

Hm, I basically use the term controller to denote any modular
component of cgroup that is not part of the cgroup core.

Do we need to be more specific/accurate than that? Especially because
there won't be any purely informational modules in cgroup2 anymore,
they'll all actually control some resource or behavior.

Re: [PATCH 1/2] cgroup: clean up the kernel configuration menu nomenclature

[Tejun Heo]

On Fri, Dec 18, 2015 at 10:23:03AM +0800, Zefan Li wrote:
> >  config CGROUP_DEBUG
> >-	bool "Example debug cgroup subsystem"
> >+	bool "Example controller"
> 
> But not all cgroup subsystems are controllers, which are debug, perf and
> cpuacct. I'm fine the other changes.

I think it's fine to call all of them controllers.  Let's use that as
the external facing noun to call all cgroup subsystems.  It is a lot
more specific and better describes the main purpose.

Thanks.

-- 
tejun

Re: [PATCH 2/2] cgroup: put controller Kconfig options in meaningful order

[Tejun Heo]

On Thu, Dec 17, 2015 at 05:19:57PM -0500, Johannes Weiner wrote:
> To make it easier to quickly find what's needed list the basic
> resource controllers of cgroup2 first - io, memory, cpu - while
> pushing the more exotic and/or legacy controllers to the bottom.
> 
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>

Applied 1-2 to cgroup/for-4.5 w/ Li's suggestion applied.

Thanks.

-- 
tejun